I am examining whether there is causal relationship between blood transfusion and infections in very small children undergoing heart surgery. My treatment groups are "high transfusion" and "low transfusion". After performing genetic matching, there remains significant imbalance for two covariates: body weight and volume of blood lost during the operation. This makes clinical sense - really small children ALWAYS require transfusion; and the greater the blood loss, the greater the transfusion. Is there anything more I can do in GenMatch so that these two covariates are balanced between the treatment groups or do I have to accept that they cannot be balanced? With thanks for any help you can offer me, Martin -- View this message in context: http://r.789695.n4.nabble.com/Genetic-Matching-GenMatch-tp4635180.html Sent from the R help mailing list archive at Nabble.com.